studentsPerT = folium.Map([38.318364, -92.412253], tiles='CartoDB Positron', zoom_start=6.5)

# generate choropleth map 
choropleth = folium.Choropleth(
    geo_data=districts,
    data=districts,
    columns=['NAME', 'StudentsPerTeacherDistrictAVG'],
    key_on='feature.properties.NAME',
    fill_color='Reds', 
    fill_opacity=1, 
    line_opacity=1,
    legend_name='Average Number of Students Per Teacher',
    highlight=True,
    smooth_factor=0).add_to(studentsPerT)

# add labels for school district and miles per school.
style_function = "font-size: 15px; font-weight: bold"
choropleth.geojson.add_child(
    folium.features.GeoJsonTooltip(['NAME', 'StudentsPerTeacherDistrictAVG'], style=style_function, labels=False))

# create a layer control
folium.LayerControl().add_to(studentsPerT)
<folium.map.LayerControl at 0x124ea15e0>
studentsPerT
Make this Notebook Trusted to load map: File -> Trust Notebook

Story

Missouri School Districts

There are two challenges that I hear a lot about when teacher and school administrator advocates come to the capitol. One is the transportation challenge that rural districts face. They have the task of transporting children from a greater land area. The other is class sizes are too large with some classes containing more than thirty students.

Looking at visualization number one, you can easily identify the more populated areas. This map is a choropleth based on the number of schools in the district. You would expect that more populated districts would have more total schools. This graphic is a good bedrock of what we 'know' before we examine the two problems mentioned earlier.

Visualization number two looks at the number of square miles for the district divided by the number of schools. This is to give perspective regarding the challenge districts may have regarding transporting children to school. Looking at those populated areas we examined earlier, we can see that this is more of an issue for rural districts with some extreme districts having schools that cover nearly a 250 square mile area.

Visualization number three looks at the district average of students per teacher. Examining this image, there is not a clear divide between rural and urban. I also noticed that districts rarely cross the ratio of 15 students per teacher. Based on this data it doesn't appear that class size is an issue if teachers are allocated equally among the district. One caveat is that I'm assuming administrators were included in the teachers number if they had a DESE teaching certificate. This most likely would deflate the teacher student average, but could add to the discussion of district efficiency.